Archive for April, 2017

Claudia Baez-Camargo

We have Claudia Baez-Camargo, Head of Governance Research, Basel Institute on Governance, giving a talk in the Social, Economic, and Decision Psychology Colloquim this Thursday (title and abstract follow).

Corruption, social norms and behaviours in East Africa: Comparative insights from Rwanda, Tanzania, and Uganda

This talk will discuss emerging findings from a comparative research project that has been investigating behavioural elements shaping the attitudes towards petty corruption among citizens in Rwanda, Tanzania, and Uganda. Preliminary analysis of the data collected suggests the salience of shared mental models about the prevalence of corruption in the communities as well as social norms of reciprocity and group obligation that contribute to making corrupt behaviours socially acceptable. These findings appear to be associated with the high levels of corruption observed in Tanzania and Uganda and with the success of the anti-corruption policies in Rwanda.

march for science

Unknown

“The March for Science champions robustly funded and publicly communicated science as a pillar of human freedom and prosperity. We unite as a diverse, nonpartisan group to call for science that upholds the common good and for political leaders and policy makers to enact evidence based policies in the public interest.”

Aging & Cognition 2017

The Aging & Cognition 2017 Conference took place in Zurich, April 20-22. There were a number of very interesting talks and posters, including very informative overview talks by Matthias Mehl, on ecological sampling to examine effects of aging in everyday life activities and cognition, and Steven Boker, on dynamic systems and distributed testing.

Baker’s report on distributed testing was particular interesting because it introduced the idea of reversing the usual experimenter-participant relation, in which the participants provide data which the experimenter analyses and archives. In Boker’s proof-of-concept study an experiment was distributed on the internet and participants were tested by interacting with a computer program on a portable device that also conducted data analysis autonomously; summary results (i.e., model likelihoods) were then automatically returned to the experimenter, such that the data was kept by the participant at all time. This type of procedure happens to have some nice statistical properties (due to model averaging) and allows participants to never give up their data (which could be particularly interesting for sensitive data about wealth or health). There are of course, some downsides to this as well, for example, one won’t necessarily be able to replicate results if participants withdraw data access at some point during or after data collection.

Of course, even in our post-factual world, scientists aren’t likely to give up control over their data anytime soon – after all, “data is the new oil”!

Marcus Lindskog

Our guest this week at the SWE colloquium is Marcus Lindskog from Uppsala University, Sweden. Marcus studies how, from infancy to adulthood, humans rely on a basic cognitive system called the Approximate Number System (ANS) and how they extract statistical information from the environment. Here are the title and abstract of his talk:

Intuitive statistics – Cognitive representations and early development

Statistical information is ubiquitous in everyday life, and one of the most important features of the human mind is its ability to extract regularities and make inferences from such data. At least since the work by Brunswik, it has been suggested that people are intuitive statisticians who spontaneously estimate statistical properties in their environment and make accurate statistical inferences. In this talk I will explore the idea of people as intuitive statisticians from a cognitive and a developmental perspective. First, I will present a series of studies investigating the cognitive processes people are engaged in when making intuitive statistical judgments from sequentially presented data. This work contrasts two models for intuitive statistical judgments, summarized in the metaphors of the lazy and eager intuitive statistician. In short, the lazy statistician postpones judgments to the time of a query when the properties of a small sample of values retrieved from memory serve as proxies for population properties. In contrast, the eager statistician abstracts summary representations of population properties online from incoming data. In the second part of the talk, I will present studies exploring intuitive statistical judgments from a developmental perspective. Recently, several studies have investigated if already infants have rudimentary abilities to summarize and make inductive inferences from small sets of data. However, very little is known about the complexity of these abilities, how they develop, or how infants process statistical information. Here, I will present a series of studies that begin addressing such questions.

Transparent science

How can we make our science more transparent? Will transparency lead to reproducibility and ultimately scientific progress? Is open science exclusively beneficial or does engaging in transparent science incur unreasonable costs for the individual researcher?

These and other questions were part of a recent lunch discussion at Cognitive and Decision Sciences. Two recent papers formed the backbone of the discussion. Munafò and colleagues (2017) offered a manifesto for reproducible science, highlighting several measures aimed at optimizing how science is conducted, presented, communicated, evaluated, and incentivized. In contrast, whilst McKiernan and colleagues (2016) advocated open science as beneficial for individual researchers’ development, career and scientific contributions, the paper also left room for debating the costs incurred by individuals engaging in transparent science. The debate amongst CDS team members spurred an interesting discussion about how much transparency is required, and identified ways in which transparency was already built into individual researchers’ projects, for instance through replication, or the sharing of analysis code and data.

The discussion will be followed up in the future with a session on where and how to preregister a scientific study. For those interested in the two sides of the reproducibility in science story, the two papers cited below provide an interesting start.

McKiernan, E. C., Bourne, P. E., Brown, C. T., Buck, S., Kenall, A., Lin, J., … & Spies, J. R. (2016). How open science helps researchers succeed. Elife, 5, e16800.

Munafò, M. R., Nosek, B. A., Bishop, D. V., Button, K. S., Chambers, C. D., du Sert, N. P., … & Ioannidis, J. P. (2017). A manifesto for reproducible science. Nature Human Behaviour, 1, 0021.

„Where to publish? Publication strategies to make your research more visible“

The University Library is organizing a course in publication strategies on Tuesday, May 2, 16.15-17.45.

Content:

You are interested in publishing your recent research or in enhancing the visibility and impact of your research? In the workshop we will outline present possibilities and address best practices of today’s electronic publishing, amongst other things pros and cons of open access publishing. We will present a few tools to help you finding the right place to publish and will make some suggestions, how your papers will be easily accessible for other researchers:

• Which journal? Which publisher? Which platform?
• Visibility and sustainability of academic publications in the digital era
• Open access: Threat or necessity for my career?
• Publishing research data?
• Support at the University of Basel: From author’s copyright to the establishment of
new journals

Target group:

For PhD students, Postdocs and other researchers of all disciplines

Requirements:

Please bring your notebooks.

Register here! until 28.04.2017

59. Tagung experimentell arbeitender Psychologen, Dresden

Die diesjährige Tagung experimentell arbeitender Psychologen (TeaP) fand vom 26.-29. März 2017 in Dresden, Deutschland, statt. Unter dem Motto “Dresden ist bunt” fand ein mindestens ebenso buntes wissenschaftliches Programm statt. Von den zahlreichen grossartigen Beiträgen möchte ich gerne auf zwei Trends, die mir besonders aufgefallen sind, eingehen.

Der erste, sehr erfreuliche, Trend, war derjenige weg von der “business as usual”-Mentalität hin zum Hinterfragen der (experimentellen und Mess-)Methoden die eingesetzt werden. So gab es eine gesonderte Talk-Session zu experimenteller Methodik, wobei vor Allem Effektstärken und Power-Analysen im Vordergrund standen. Im Symposium zu kognitiver Modellierung wurden verschiedene Methoden der Modellevaluation und des Modellvergleichs vorgestellt und teilweise gegeneinander verglichen. Aber auch in anderen, allgemeineren Sessionen waren methodische Beiträge zu finden.

Der zweite Trend war derjenige zur Dezentralisierung der Datenerhebung. Felix Henninger hat etwa die neue Experimentalsoftware lab.js vorgestellt. Diese Software erlaubt es, schnell und einfach psychologische Experimente zu programmieren, die sich dann im Web-Browser ausführen lassen. Doch nicht nur der Zugang zu Experimentalsoftware wird zunehmend einfacher, auch die technischen Möglichkeiten und psychometrische Qualitäten von Online-Experimenten werden zunehmend besser. Kilian Semmelmann etwa macht Forschung zur Messgenauigkeit von Antwortlatenzen bei browserbasierten Experimenten (fast keine Unterschiede zu gängiger Experimentalsoftware) oder auch zur Messung von Blickbewegungsdaten mithilfe der Webcam (was erstaunlich präzise zu funktionieren scheint).

Aus den SWE-Arbeitsgruppen hat Dirk Wulff Vorträge im Symposium zu kognitiver Modellierung (über verschiedene Modellvergleichsmethoden) sowie im Symposium zur Messung von Blick- und Mauscursorbewegungen (über verschiedene Ebenen der Analyse von Mauscursorbewegungen und potenzielle Aggregationsartefakte) gehalten. Ich habe die Resultate eines Projekts vorgestellt, bei dem wir Verhaltens- und Blickbewegungsdaten analysiert sowie kognitive Modellierung eingesetzt haben, um den Einfluss von Aufmerksamkeit auf Entscheidungen zu untersuchen.

Insgesamt war es eine sehr interessante TeaP und ich freue mich auf die 60. TeaP, die nächstes Jahr in Marburg stattfinden wird!

Christopher Summerfield

Our guest this week at the SWE colloquium is Christopher Summerfield from Oxford University. Christopher studies both perceptual and value-based decision making and combines mathematical modeling with neuroscientific approaches. Here are the title and abstract of his talk:

Optimality and “irrationality” in human decision making

Humans make “near-optimal” category judgments about noisy sensory stimuli, but on cognitive tasks they often exhibit systematic biases that fail to maximize economic outcomes. In my talk, I will discuss why. I will argue that because the “ideal” observer framework considers only noise that arises during sensory encoding, it frequently misspecifies the decision policy that will maximize rewards. When we also consider “late” noise (that arising during information integration) cognitive biases can often be reframed as efficient, reward-maximizing policies. I will discuss with reference to data and modeling from tasks involving perceptual averaging, transitive choices and decoy effects in multialternative choices.

Sozial-, Wirtschafts- und Entscheidungspsychologie

Social, Economic, and Decision Psychology Blog